Additionally, the lake centre is deeper than other parts of the lake and has a mineral soil due to the ongoing resuspension ( Fig. 8) which has prevented the development of an organic-rich lacustrine sediment ( Shen

et al., 2011). Due to unsuitability of the lake centre for macrophyte growth, alternative stable states are most likely not present here. The large variation in nutrients and suspended solids indicate a low internal connectivity in Taihu, especially between the east and the rest of the lake ( Li et al., SCH772984 2011a). Likely, the variation in concentrations is the result of the long residence time of 300 days. The positive effect of low connectivity on water quality in East Taihu is enhanced by the prevailing winds that blow floating algae away from

the east ( Li et al., 2011b and Qin et al., 2010). If the flushing rate in the lake would be higher, nutrient concentrations would most likely be more equally distributed and macrophytes in the east would be more affected by eutrophication ( Fig. 2C, process 6). A whole-lake flushing measure to reduce water age in Taihu ( Li et al., 2011b) revealed the effect of internal connectivity on Taihu’s water quality. The water age reduction was meant to flush out nuisance algae. However, the water age could not be shortened enough to overcome cyanobacterial growth ( Qin et al., 2010). At the same time the most eutrophic bay (Meiliang) did not significantly improve in water quality as a result of flushing because local prevailing currents prevented inflow of the water into the bay ( Li et al., 2013 and Qin et al., 2010). In the past, the internal connectivity of this bay has been

Talazoparib datasheet decreased by land reclamation; some islands became peninsula and thereby separated bays that where connected before ( Hu et al., 2004 and Li, 1999). Hence, the low internal connectivity Vildagliptin prevents exchange between bays and restricts propagation of the phytoplankton towards the east where at present macrophytes still prevail. Consequently, Taihu has a modular response to eutrophication leading to different states side by side, some states conceivably alternatively stable, others probably not. It is interesting to know whether lake size, spatial heterogeneity and internal connectivity acting in Taihu are exemplary for the existence and patterning of alternative stable states in other large shallow lakes. Based on a number of large shallow lakes listed in Table 1, the generality of these mechanisms will be discussed. The estimated probability of other large shallow lakes to have macrophyte domination, and, if so, whether these have a potentially alternative stable state is shown in Fig. 9A. A comparison with the model outcomes indicate that 8 lakes might have alternative stable states (Table 1) including Lake Apopka (USA) where literature previously presumed alternative stable states (Bachmann et al., 1999 and Lowe et al., 2001).